Utilization of primary and secondary medical care among disadvantaged populations: a log-linear model analysis

Gregory Y.om Din, Zinaida Zugman, Alla Khashper

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

AIM: We examined how, where an overall population is covered by universal health insurance, characteristics of disadvantaged populations interact to influence inequality in primary and secondary medical care utilization.

SUBJECTS & METHODS: Disadvantaged populations, the focus of the study, were defined as populations who have lower socio-economic status (SES), who are elderly and/or reside in a peripheral area. Data from the 2009 Israeli National Health Survey were analysed using log-linear models to estimate utilization of medical care.

RESULTS: The main findings were: a) pro-poor utilization of primary medical care among elderly populations, with higher odds ratios for low SES populations in the periphery; (b) lack of interaction between SES and primary medical care utilization among younger populations, between SES and secondary medical care utilization among the elderly and pro-rich utilization of secondary medical care among younger populations who did not regularly visit general practitioners (GP); (c) the odds ratios of secondary medical care utilization increased as SES decreased for both elderly and younger populations who also regularly visited a GP.

CONCLUSION: Potential policy implications for disadvantaged populations, regarding possible inequality in primary and secondary medical care utilization, can be drawn using log-linear model analysis of interactions among characteristics (SES, age, location) of disadvantaged populations.

Original languageEnglish
Pages (from-to)9-21
Number of pages13
JournalGlobal journal of health science
Volume6
Issue number5
DOIs
StatePublished - 1 Sep 2014
Externally publishedYes

Fingerprint

Dive into the research topics of 'Utilization of primary and secondary medical care among disadvantaged populations: a log-linear model analysis'. Together they form a unique fingerprint.

Cite this